Mechanistic studies of CO2 reduction to methanol mediated by an N-heterocyclic germylene hydride

Article abstract of DOI:10.1039/c3dt53321b

The labile germylene hydride L^CyGeH is capable of activating CO₂ affording the corresponding formate L^CyGeOCH(O) (2) (L^Cy = cyclo-C₆H₈-1-NAr-2-C(Ph)NAr, Ar = 2,6-iPr₂C₆H₃). Compound 2 and the previously reported LGeOCH(O) (L = CH(MeCNAr)₂, Ar = 2,6-iPr₂C ₆H₃) (2′) could be further converted to methanol with the AlH₃·NMe₃ alane-amine adduct as a hydrogen source upon workup with water. A plausible mechanism for the conversion of the formate complexes to methanol is proposed based on additional results from the conversion of 2′ with the milder hydride delivery agent LAlH₂. This journal is the Partner Organisations 2014.

Full text of DOI:10.1039/c3dt53321b

Volume 43 Number 16 28 April 2014 Pages 5937–6270
Dalton
Transactions
An international journal of inorganic chemistry
www.rsc.org/dalton
ISSN 1477-9226
COVER ARTICLE
Driess et al.
Mechanistic studies of CO2 reduction to methanol mediated by an
N-heterocyclic germylene hydride

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Mechanistic studies of CO2 reduction to
methanol mediated by an N-heterocyclic
germylene hydride†
Cite this: Dalton Trans., 2014, 43,
6006
Gengwen Tan,a Wenyuan Wang,a,b Burgert Bloma and Matthias Driess*a
The labile germylene hydride LCyGeH is capable of activating CO2 affording the corresponding formate
L
CyGeOCH(vO) (2) (LCy = cyclo-C6H8-1-NAr-2-C(Ph)NAr, Ar = 2,6-iPr2C6H3). Compound 2 and the
Received 25th November 2013,
Accepted 2nd January 2014
previously reported LGeOCH(vO) (L = CH(MeCvNAr)2, Ar = 2,6-iPr2C6H3) (2’) could be further
converted to methanol with the AlH3·NMe3 alane-amine adduct as a hydrogen source upon workup with
water. A plausible mechanism for the conversion of the formate complexes to methanol is proposed
based on additional results from the conversion of 2’ with the milder hydride delivery agent LAlH2.
DOI: 10.1039/c3dt53321b
www.rsc.org/dalton
intermediates were isolated in this study. Very recently, a
theoretical study by Sakaki et al. has shown that LGeH could
Introduction
The activation and conversion of carbon dioxide to valuable also act as a catalyst for CO2 hydrosilylation to F3SiOCH(vO)
chemicals are attracting increasing attention in chemical when HSiF3 is used as the hydride source.10 Until now, there
research due to the energy and climate crisis.1 Hydrogenation have been only two examples of using the germylene hydride
of carbon dioxide to other C1 feedstocks, such as formic acid, complexes bearing β-diketiminato ligands for CO2 activation,11
methanol and methane, is one of the most straightforward so it seems desirable to apply varied ligand scaffolds to stabil-
approaches for utilising CO2.2 The process, which is catalysed ise the highly active germylene hydride species and investigate
by transition metal complexes, has been significantly develo- their ability for CO2 activation.
ped in the last decade.2,3 Small molecule activation represents
In 2010, we introduced the 2-iminocyclohexylidenebenzyl-
one of the most crucial research topics in contemporary main amine ligand LCyH (LCy = cyclo-C6H8-1-NAr-2-C(Ph)NAr, Ar = 2,6-
group chemistry.4 However, examples of CO2 hydrogenation iPr2C6H3) for stabilising a germylene complex, and studied its
mediated by main-group pre-catalysts are still scarce. Recently, reactivity towards water and ammonia.12 Since the successful
N-heterocyclic carbenes (NHCs) demonstrated by Ying et al.5 application of this ligand, we wanted to apply it also as a sup-
and ‘Frustrated Lewis Pair’ (FLP) systems reported by Stephan6 porting ligand for the corresponding germylene hydride and
and Piers7 were shown to serve as suitable hydrogenation the potentially isolable corresponding formate complexes.
systems. In line with that, Roesky et al. have demonstrated that Herein, we report the use of this ligand (LCyH) for the syn-
the germylene hydride LGeH (L = CH(MeCvNAr)2, Ar = 2, thesis of a new germylene hydride derivative and its appli-
6-iPr2C6H3) can activate CO2 affording a germylene formate cation in hydrogenation of CO2 to the germylene formate
complex LGeOCH(vO) (2′),8 which can be further converted to LCyGeOCH(vO) (2). Complex 2 and the reported LGeOCH
formic acid and methanol upon hydrolysis with water using (vO) (2′) react readily with alane (AlH3·NMe3) to afford deuter-
LiNH2·BH3 and NH3·BH3 adducts as hydride sources.9 In ated methanol (CH3OD) upon hydrolysis with D2O. The trap-
addition, they proposed a possible mechanism for the latter ping and full elucidation of key intermediates in the
case based on NMR spectroscopic investigations, but no conversion of 2′ to methanol are also reported and shed light
on the mechanism of CO2 reduction mediated by germylene
hydrides.
aTechnische Universität Berlin, Department of Chemistry: Metalorganics and
Inorganic Materials, Sekr. C2, Strasse des 17. Juni 135, 10623 Berlin, Germany.
E-mail: Matthias.driess@tu-berlin.de; Fax: (+49)30-314-29732
bCollege of Chemistry & Materials Science, Northwest University, Xi’an 710069,
Results and discussion
P.R. China
†Electronic supplementary information (ESI) available: Experimental details
Accordingly, the synthesis of the LCyH stabilised germylene
and crystal data and structure refinement for 1–4. CCDC 967657–967660. For
hydride complex was first investigated. Akin to the preparation
of LGeH,8 LCyGeCl was treated with one molar equivalent of
ESI and crystallographic data in CIF or other electronic format see DOI:
10.1039/c3dt53321b
6006 | Dalton Trans., 2014, 43, 6006–6011
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Scheme 1 Synthesis of compound 1 via Ge(II) hydride.
K[BH(sBu)3] in toluene at −78 °C and allowed to warm to room
temperature. Unexpectedly, after a reaction time of 12 hours at
ambient temperature, the expected germylene hydride
complex (LCyGeH) was not isolated. Instead, LCy(H)Ge (1) was
isolated as a single product with 84% yield (Scheme 1).
However, when the reaction process was monitored by
1H NMR spectroscopy, after the mixture was stirred for 3 hours
at room temperature, the characteristic resonance signal for
Ge–H proton could indeed be observed at δ = 7.96 ppm which
is comparable to that of LGeH (δ = 8.08 ppm).11a This indicates
that LCyGeH is formed during the reaction, but it is labile
and ultimately undergoes a 1,3-hydrogen transfer from the
germanium centre to the backbone of the ligand to give the
germylene 1 with a hydrogenated ligand scaffold as the
thermodynamic product. A similar reaction mode was reported
by Jones et al. when they attempted to use tBuNacnacH (tBuNac-
nac = CH(tBuCNAr)2, Ar = 2,6-iPr2C6H3) and tBuMesNacnacH
ligands
( = CH(tBuCNMes)2, Mes = 2,4,6-
tBuMesNacnac
Me3C6H3) to stabilise a germylene hydride. In both cases the
only isolated products are diamido germylene complexes with
1,3-hydride migration to the backbone of the ligands.11b,13
However, compound 1 cannot activate CO2 even at elevated
temperatures in toluene or benzene.
Fig. 1 Molecular structures of compounds 1 (a) and 2 (b) in the solid
state. Thermal ellipsoids are drawn at the 50% probability level; the Dipp
groups are depicted in wireframe style. Disorders at C36 and C38 of
compound 2 and hydrogen atoms (except those at C19 of compound 1
and C37 of compound 2) are omitted for clarity.
Although LCyGeH is only a kinetic product, we reasoned
that it could be employed for hydrogenation of CO2 to afford
the germylene formate complex LCyGeOCH(vO) (2). Following
this idea, LCyGeCl was reacted with K[BH(sBu)3] in toluene at
−5 °C for 12 hours under a N2 atmosphere, and then the gas
atmosphere was changed with CO2 through a freeze–pump–
thaw cycle, and the reaction mixture was further stirred for five
hours. From the reaction mixture, the germylene formate
LCyGeOCH(vO) (2) could indeed be isolated in 77% yield
(Scheme 2).
Both compounds 1 and 2 are yellowish solids, and are
thermally robust (M.p. 178 °C (1); 165 °C (2)) without any
decomposition when stored under a N2 atmosphere at room
temperature for several months. They are soluble in toluene,
benzene and THF, and slightly soluble in n-hexane. They were
fully characterised by NMR spectroscopy (1H, 13C), mass spec-
trometry, elemental analyses as well as single crystal X-ray
diffraction analyses. The resonance signal for the γ-H proton
(PhCHNAr) in complex 1 is observed at δ = 4.75 ppm as a
singlet in the 1H NMR spectrum in C6D6 at room temperature.
The corresponding carbon nucleus resonates at δ = 74.8 ppm
in the 13C{1H} NMR spectrum. In the APCI-HR-MS spectrum,
the signal for the molecular ion peak of 1 is found at m/z
595.3082 (calcd: m/z 595.3102), whereas the molecular ion
peak for compound 2 is not found, but the signal for the mole-
cule fragment corresponding to loss of the formate group is
observed at m/z 593.2936 (calcd: m/z 593.2946). The resonance
signal for the proton at the formate group in 2 is shown at δ =
8.78 ppm in the 1H NMR spectrum, which is comparable to
that of the formate group in 2′ (δ = 8.4 ppm); the corres-
ponding 13C signal is revealed at δ = 164.4 ppm in the
13C NMR spectrum. The molecular structures of complexes 1
and 2 are shown in Fig. 1.
Complex 1 crystallises in the triclinic space group P1, which
is a chiral crystal system. The germanium centre is coordinated
by the N1 and N2 atoms. The distances of Ge1 to N1 and N2
are 1.8250(19) Å and 1.867(2) Å, respectively. They are compar-
able to those in LCy′Ge (LCy′ = cyclo-C6H7-1-NAr-2-C(Ph)NAr,
Ar = 2,6-iPr2C6H3) which bears a dianionic ligand (1.861 and
1.843 Å).12 The bond distances of C14–C19 (1.513 (3) Å) and
Scheme 2 Synthesis of compound 2.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 6006–6011 | 6007

View Article Online
Paper
Dalton Transactions
C13–C14 (1.346(3) Å) indicate the single and double bond
character, respectively. This is in accordance with the structure
portrayed for 1 in Scheme 1. Compound 2 crystallises in
the monoclinic space group P2/c and the germanium centre
features a trigonal pyramidal geometry. The Ge1 centre devi-
ates from the plane defined by N1–C2–C3–C4–N2 by 0.534 Å.
The distances of Ge1–N1, Ge1–N2 (1.976(2) and 1.988(2) Å) and
Ge1–O1 (1.958(2) Å) are consistent with those observed in 2′
(Ge–N: 1.969(2) and 1.968(2) Å, Ge–O: 1.9339(18) Å).8
Roesky et al. demonstrated that the germylene formate 2′
could be converted to formic acid and methanol with
LiNH2·BH3 and NH3·BH3 as the hydride sources upon workup Scheme 3 Syntheses of compounds 3 and 4.
with water, respectively.9 Very recently, Sakaki et al. showed
that LGeH can act as a catalyst in CO2 hydrogenation when a
formate group to the OCH2O group represents the second step
for CO2 hydrogenation. While there is still one Al–H in com-
pound 3, the yellow solution of 3 in THF gradually turned to
red at room temperature, and so we tested the thermal stability
of 3. From a 1H NMR spectroscopic investigation, compound 3
was decomposed to give LGeH and compound 4 after heating
at 60 °C for 3 hours in THF (Scheme 3). From the reaction
mixture, compound 4 was isolated as a colorless crystalline
product in 60% yield. The proton signals for the OCH2O group
and the γ-H proton in the β-diketiminate ligand are observed
at δ = 4.74 and 5.00 ppm, respectively, with the integrate ratio
of 4 : 2. In this step, the LGeH is regenerated. Moreover, we
tried to cleave the O–CH2O bond in 4 with various hydride
sources, but all attempts failed, probably due to the steric
crowding resulting from the bulky β-diketiminato ligand.
The molecular structures of compounds 3 and 4 are shown
in Fig. 3 and 4. Complexes 3 and 4 crystallise in the mono-
clinic space group C2/c. The Ge1 has a trigonal pyramidal geo-
metry and Al1 feature a tetrahedral coordination in 3, and they
are bridged by the OCH2O group. The Ge1–O1 bond length
(1.854(4) Å) is shorter than that in the starting material 2′
(1.9339(18) Å),8 whereas it is comparable to that in LGeOiPr
suitable silane is applied as the hydride source based on a
theoretical study.10 Inspired by these results, we were inter-
ested in introducing alane as a hydride transfer source for CO2
hydrogenation and elucidating the mechanism of the reaction.
We chose Me3N·AlH3 as the hydride source and used it to
react with 1/3 molar equivalent of 2′ and compound 2, respect-
1
ively. The H NMR spectra in C6D6 showed that both the reac-
tions proceeded smoothly to give the corresponding germylene
hydride complexes in almost quantitatively yields within one
hour at room temperature (Fig. 2 and ESI†).14 After stirring for
two hours in toluene, the reactions were quenched with D2O at
0 °C. The yields of CH3OD were determined as 46% (2′) and
42% (2) by 1H NMR spectroscopy with 1,4-dioxane as an
internal standard.
After obtaining these results, we proceeded to elucidate the
stepwise process of this reaction. For this purpose, LAlH2 (L =
CH(MeCvNAr2)2, Ar = 2,6-iPr2C6H3)15 was applied as a milder
hydride delivery agent to react with 2′. By the reaction of 2′
with one molar equivalent of LAlH2, the carbonyl group is
further hydrogenated to afford the striking compound 3 as an
OCH2O bridged heterobimetallic complex (Scheme 3). The
resonance signal for the protons at the OCH2O group is observed
at δ = 4.48 ppm, the γ-H protons of the β-diketiminato ligands
resonate at δ = 4.90 and 5.02 ppm, respectively, and their inte-
grate ratio is 2 : 1 : 1, which is in accordance with the structure
depicted for compound 3. Hence, the hydrogenation of the
Fig. 3 Molecular structure of compound 3 in the solid state. Thermal
ellipsoids are drawn at the 50% probability level; the Dipp groups are
depicted in wireframe style. Disorders at the core part GeOCH2OAl of
compound 3 and hydrogen atoms (except those at C30 and Al1) are
omitted for clarity. Operation symmetry for all atoms labelled with “A”:
−x + 1/2,−y + 1/2,−z + 2 (3).
Fig. 2 1H NMR spectra of the reaction of 2’ with 3 molar equivalents of
NMe3·AlH3 for 60 min (top) and 2’ in C6D6 (bottom).
6008 | Dalton Trans., 2014, 43, 6006–6011
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
water. The germylene hydrides are concomitantly regenerated
in this process. Based on the reaction of the mild hydrogen
delivery agent (LAlH2) with LGeOCH(vO) (2′) as a model reac-
tion, we suggested a plausible mechanism for the formation of
methanol with the isolation of LGeOCH2OAl(H)L (3) and
(LAlOCH2O)2 (4), respectively. These studies shed new light on
the reduction of CO2 mediated by germylene hydrides.
Experimental section
All experiments were carried out under dry oxygen-free nitro-
gen using standard Schlenk techniques. Solvents were dried by
standard methods and freshly distilled and degassed prior to
use. The NMR spectra were recorded on Bruker spectrometers
Fig. 4 Molecular structure of compound 4 in the solid state. Thermal
ellipsoids are drawn at the 50% probability level; the Dipp groups are
depicted in wireframe style. Hydrogen atoms (except those at C30 and (AV400 or AV200) referenced to residual solvent signals as
internal standards (1H NMR: CDCl3, 7.26 ppm; C6D6,
C30A) are omitted for clarity. Operation symmetry for all atoms labelled
with “A”: −x + 1/2,−y + 1/2,−z + 1 (4).
7.16 ppm; and 13C{H} NMR: CDCl3, 77.0 ppm; C6D6,
128.1 ppm) or with an external standard. Concentrated solu-
tions of samples in C6D6 or CDCl3 were sealed off in a Young-
type NMR tube for measurements. Melting points were
recorded on a “Melting point tester” device from BSGT
Company and are uncorrected. All the samples were sealed off
in capillary under vacuum and each sample was measured in
duplicate. High resolution mass spectra (APCI, atmosphere pressure
chemical ionization) were recorded on an Orbitrap LTQ XL of a
Thermo Scientific mass spectrometer and the raw data were
evaluated using the Xcalibur computer program. For the single
crystal X-ray structure analyses the crystals were each mounted
on a glass capillary in perfluorinated oil and measured in a
cold N2 flow. The data of compounds 1 and 2 were collected
on an Oxford Diffraction Xcalibur S Sapphire at 150 K (Mo-Kα
radiation, λ = 0.71073 Å), and the data of compounds 3 and 4
were collected on an Oxford Diffraction Supernova, Single
source at offset, Atlas at 150 K (Cu-Kα radiation, λ = 1.5418 Å).
The structures were solved by direct methods and refined on
F2 with the SHELX-9719 software package. The positions of the
H atoms were calculated and considered isotropically accord-
ing to a riding model.
Scheme 4 Plausible reaction mechanism for germylene mediated CO2
reduction to CH3OD with alane.
(1.821(2) Å).16 The Al1–O2 bond distance (1.807(3) Å) is akin to
those in [{LAlMe(μ-O)AlMe2}]2 (av. 1.8493 Å).17 Compound 4 is
a binuclear aluminium complex with a (AlOCH2O)2 core struc-
ture. The Al–N (1.8979(13) and 1.8963(13) Å) and Al–O (1.7123
(11) and 1.7239(11) Å) bond lengths are similar to those in LAl-
[OB(3-MeC6H4)]2(μ-O) (Al–N: 1.872(2) and 1.862(2) Å; Al–O:
1.7362(17) and 1.7418(17) Å).18
Based on the model reaction of LAlH2 with 2′ and the iso-
lation of complexes 3 and 4, the stepwise process of the con-
version of 2 or 2′ with Me3N·AlH3 to methanol upon hydrolysis
with water could be explained as follows (Scheme 4): complex
2′ is hydrogenated to form the adduct LGeOCH2OAlH2·NMe3,
and the subsequent Al–H hydride of Me3N·AlH3 transfer to the
Ge centre regenerates LGeH. Concomitantly, Me3N·AlH2O-
CH2OAlH2·NMe3 is formed which is continuously converted to
Me3N·AlH2–OMe and Me3N·AlH2OH2Al·NMe3 with O–CH2O
bond cleavage. Hydrolysis of Me3N·AlH2–OMe with D2O then
yields CH3OD as a C1 product.
Commercially available reagents were purchased from
Aldrich, Acros and used as received. LCyH, LCyGeCl (LCy
=
cyclo-C6H8-1-NAr-2-C(Ph)NAr, Ar = 2,6-iPr2C6H3),12 LGeOCH-
15
(vO) (2′)8 and LAlH2 (L = CH[C(Me)NAr]2, Ar = 2,6-iPr2C6H3)
were synthesized according to published procedures.
Compound LCy(H)Ge (1)
L
CyGeCl (0.628 g, 1 mmol) was placed in a Schlenk flask
(100 mL) in the glovebox. Toluene (30 mL) was transferred into
the flask via cannula under stirring at room temperature and a
clear yellow solution was formed. The solution was cooled to
−78 °C, and K[BH(sBu)3] (1 mL, 1 mmol, 1 M solution in THF)
was added dropwise to the solution via a syringe. The mixture
Conclusion
In conclusion, we demonstrated that the novel formate com- was allowed to warm to room temperature and stirred for
pound LCyGeOCH(vO) (2) and the previously reported another 12 hours. The obtained red solution was concentrated
LGeOCH(vO) (2′) can both be efficiently hydrogenated to yield to 10 mL and filtered. The filtrate was left at 0 °C for 24 hours
methanol using alane as the hydride source upon workup with to afford a yellow crystalline product (1). The product was
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 6006–6011 | 6009

View Article Online
Paper
Dalton Transactions
collected by decantation of the supernatant and the obtained calcd for [C37H47GeN2 (M
−
CO2H)]+: m/z 593.2946;
solid was dried in vacuo for several hours. Yield: 0.50 g found: m/z 593.2936. IR (KBr):
v = 2870 (OC(vO)−H),
(0.84 mmol, 84%). M.p. 178 °C (dec.). 1H NMR (200.1 MHz, 1657(OC(vO)−H) cm−1
.
3
C6D6, 298 K): δ = 0.53 (d, 3 H, JH–H = 6.8 Hz, CH(CH3)2), 1.10
Compound LGeOCH2OAl(H)L (3)
(d, 3 H, 3JH–H = 7.0 Hz, CH(CH3)2), 1.25 (d, 3 H, 3JH–H = 6.6 Hz,
3
CH(CH3)2), 1.28 (d, 9 H, JH–H = 6.8 Hz, CH(CH3)2), 1.37–1.44 LGeOCH(vO) (2′) (0.268 g, 0.5 mmol) and LAlH2 (0.224 g,
3
(m, 7 H, Cy-CH2 (4 H) + CH(CH3)2 (3 H)), 1.47 (d, 3 H, JH–H
=
0.5 mmol) were placed in a Schlenk flask (100 mL) in the
6.8 Hz, CH(CH3)2), 1.75–2.01 (m, 4 H, Cy-CH2), 3.13 (sept, 1 H, glovebox. Toluene (30 mL) was transferred to the mixture via
3
3JH–H = 6.8 Hz, CH(CH3)2), 3.55 (sept, 1 H, JH–H = 6.8 Hz, CH- cannula under stirring at −50 °C. The solution was allowed to
(CH3)2), 3.91 (sept, 2 H, 3JH–H = 7.0 Hz, CH(CH3)2), 4.75 (s, 1 H, warm to room temperature and stirred for another 12 hours to
γ-H (PhCHNAr)), 7.04–7.41 (m, 11 H, Ar-H and Ph-H). 13C{1H} give a clear yellow solution. The solution was concentrated to
NMR (50.3 MHz, C6D6, 298 K): δ = 22.5, 22.9 (iPr-CH3), 23.0 ca. 5 mL and filtrated. The obtained filtrate was left at 0 °C for
(Cy-CH2), 23.5(iPr-CH3), 23.7 (Cy-CH2), 23.8 (iPr-CH3), 25.2, 24 hours to give yellow crystals of compound 3. The product
26.1, 26.2, 26.9 (iPr-CH), 27.9 (Cy-CH2), 28.1, 28.2, 28.3, 28.4 was collected by decantation of the supernatant and dried
(iPr-CH3), 29.9 (Cy-CH2), 74.8 (γ-C), 109.1 (CH2CCNAr), 123.2, in vacuo for several hours. Yield: 0.36 g (0.37 mmol, 74%).
123.8, 124.0, 124.2, 126.4, 127.0, 127.1, 128.2, 128.3 (Ar-CH), M.p. 96 °C (dec.). 1H NMR (400.2 MHz, CDCl3, 298 K): δ = 0.62
134.4, 140.3, 143.7, 145.0, 146.9, 147.1, 147.2, 148.1 (Ar-C and (d, 6 H, 3JH–H = 6.4 Hz, CH(CH3)2), 0.79 (d, 6 H, 3JH–H = 6.8 Hz,
CH2CCNAr). Elemental analysis for C37H48GeN2: calcd: C, CH(CH3)2), 0.94 (d, 6 H, 3JH–H = 6.8 Hz, CH(CH3)2), 1.01 (d, 6 H,
3
74.89; N, 4.72; H, 8.15; found: C, 74.19; N, 4.48; H, 8.30. 3JH–H = 6.8 Hz, CH(CH3)2), 1.02 (d, 6 H, JH–H = 6.8 Hz,
APCI-HR-MS: calcd for [C37H49GeN2 (M + H)]+: m/z 595.3102; CH(CH3)2), 1.07 (d, 6 H, 3JH–H = 6.8 Hz, CH(CH3)2), 1.15 (d, 6 H,
3
found: m/z 595.3082.
3JH–H = 6.8 Hz, CH(CH3)2), 1.21 (d, 6 H, JH–H = 6.8 Hz,
CH(CH3)2), 1.62 (s, 6 H, α-CH3), 1.67 (s, 6 H, α-CH3), 2.98–3.12 (m,
6 H, CH(CH3)2), 3.17 (sept, 2 H, 3JH–H = 6.8 Hz, CH(CH3)2), 4.48
Compound LCyGeOCH(vO) (2)
L
CyGeCl (1.26 g, 2 mmol) was placed in a Schlenk flask (s, 2 H, OCH2O), 4.90 (s, 1 H, γ-H), 5.02 (s, 1 H, γ-H), 6.84–6.86
(100 mL) in the glovebox. Toluene (30 mL) was transferred into (m, 2 H, Ar-H), 6.98–7.03 (m, 2 H, Ar-H), 7.06–7.26 (m, 8 H,
the flask via cannula under stirring at room temperature and a Ar-H). The resonance signal for Al-H is not observed in the spec-
clear yellow solution was formed. The solution was cooled to trum. 13C{1H} NMR (100.1 Hz, CDCl3, 298 K): δ = 23.1 (α-CH3),
−78 °C, and K[BH(sBu)3] (2 mL, 2 mmol, 1 M solution in THF) 23.4(α-CH3), 24.1, 24.2, 24.6, 24.7, 25.1, 25.8 (iPr-CH3), 27.3,
was added dropwise to the solution via a syringe. The mixture 27.7, 28.0, 28.7 (iPr-CH), 85.9 (OCH2O), 96.2 (γ-C), 96.4 (γ-C),
was then placed in a cooled water–salt bath (ca. −5 °C), and 123.3, 123.9, 124.1, 124.8, 126.3, 126.7, 126.8, 128.2, 129.0 (Ar-
allowed to stir at this temperature for 12 hours. Then the CH), 139.4, 140.7, 143.1, 143.8, 143.9, 146.0 (Ar-C), 163.2
atmosphere was changed to CO2 by a freeze–pump–thaw cycle, (CNAr), 169.9 (CNAr). Elemental analysis for C59H85AlGeN4O2
and the mixture was stirred under a CO2 atmosphere for (%): calcd: C, 72.17; N, 5.71; H, 8.72; found: C, 72.40; N, 5.86;
12 hours. All the volatiles were removed in vacuo and the H, 8.55. APCI-HR-MS: calcd for [C59H86AlGeN4O2 (M + H)]+:
residue was washed with hot (ca. 50 °C) n-hexane (20 mL), and m/z 983.5772; found: m/z 983.5751.
the solution was filtrated. The remaining residue is the pure
product (2) on the basis of H NMR spectroscopy. The filtrate
1
Compound LAl(OCH2O)2AlL (4)
was left at room temperature for 12 hours to yield crystals of 2, Compound 3 (0.491 g, 0.5 mmol) was placed in a Schlenk flask
which are suitable for X-ray single crystal diffraction analysis. (50 mL) in the glovebox. THF (10 mL) was transferred to the
Total yield: 1.01 g (1.55 mmol, 77%). M.p. 165 °C (dec.). flask via cannula at room temperature. The yellow solution was
3
1H NMR (200.1 MHz, C6D6, 298 K): δ = 0.91 (d, 3 H, JH–H
=
heated at 60 °C for 12 hours under stirring, and an orange-red
3
6.6 Hz, CH(CH3)2), 1.02 (d, 3 H, JH–H = 6.8 Hz, CH(CH3)2), solution was formed. All volatiles were removed in vacuo. The
1.12–1.23 (m, 16 H, Cy-H (4 H) + CH(CH3)2 (12 H)), 1.27 (d, residue was extracted firstly with n-hexane (10 mL) (to remove
3
3
3 H, JH–H = 6.6 Hz, CH(CH3)2), 1.33 (d, 3 H, JH–H = 6.6 Hz, the LGeH), and then it was extracted with toluene (10 mL) to
CH(CH3)2), 1.86–2.21 (m, 4 H, Cy-H), 3.03–3.31 (m, 2 H, give a yellow filtrate. The toluene solution was concentrated to
CH(CH3)2), 3.44–3.72 (m, 2 H, CH(CH3)2), 6.67–6.92 (m, 7 H, Ar-H ca. 5 mL and filtrated. The filtrate was left at 0 °C for 24 hours
and Ph-H), 7.03–7.22 (m, 4 H, Ar-H and Ph-H), 8.78 (s, 1 H, to give colorless crystals of compound 4. The product was col-
–OCH(vO)). 13C{1H} NMR (50.3 MHz, C6D6, 298 K): δ = 21.0 lected by removing the mother liquor and dried under vacuum
(Cy-CH2), 22.3(Cy-CH2), 22.9, 23.3, 24.2, 24.3, 24.4, 26.8 for several hours. The mother liquor was further concentrated
(iPr-CH3), 27.2, 27.6 (iPr-CH), 27.9 (iPr-CH3), 28.6, 28.7(iPr-CH), to ca. 3 mL, and afforded another portion of product after crys-
28.8 (Cy-CH2), 28.9(iPr-CH3), 31.1(Cy-CH2), 107.0 (γ-C), 123.2, tallization at −30 °C. Total yield: 0.15 g (0.15 mmol, 60%).
124.2, 125.3, 126.9, 127.0, 127.5, 127.7, 127.8, 127.9, 128.0, M.p. 191 °C. 1H NMR (200.1 MHz, CDCl3, 298 K): δ = 0.78 (d, 24
3
3
128.6 (Ar-CH), 137.9, 139.9, 140.4, 143.0, 145.1, 146.0, 146.7 H, JH–H = 6.6 Hz, CH(CH3)2), 0.95 (d, 24 H, JH–H = 6.8 Hz,
(Ar-C), 164.4 (–OCH(vO)), 165.8 (Cy-CN), 168.7 (Ph-CN). CH(CH3)2), 1.55 (s, 12 H, α-CH3), 3.06 (sept, 8 H, 3JH–H = 6.8 Hz,
Elemental analysis for C38H48GeN2O2 (%): calcd: C, 71.60; N, CH(CH3)2), 4.74 (s, 4 H, OCH2O), 5.00 (s, 2 H, γ-H), 6.96 (s, 3 H,
4.39; H, 7.59; found: C, 71.56; N, 4.37; H, 7.93.APCI-HR-MS: Ar-H), 7.00 (s, 4 H, Ar-H), 7.14–7.21 (m, 5 H, Ar-H). 13C{1H}
6010 | Dalton Trans., 2014, 43, 6006–6011
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
NMR (50.3 Hz, CDCl3, 298 K): δ = 23.5 (α-CH3), 24.4(CH-
(CH3)2), 24.7(CH(CH3)2), 27.9 (CH(CH3)2), 85.7 (OCH2O), 97.1
(γ-C), 124.1, 126.8 (Ar-CH), 140.4, 144.2 (Ar-C), 170.3 (CNAr).
Elemental analysis for C60H86Al2N4O4 (%): calcd: C, 73.44; N,
5.71; H, 8.83; found: C, 73.72; N, 5.89; H, 8.65. APCI-HR-MS:
calcd for [C60H87Al2N4O4 (M + H)]+: m/z 981.6353; found: m/z
981.6340.
A. Spannenberg and M. Beller, J. Am. Chem. Soc., 2012, 134,
20701.
4 S. Yao, Y. Xiong and M. Driess, Organometallics, 2011, 30,
1748.
5 S. N. Riduan, Y. Zhang and J. Y. Ying, Angew. Chem., Int.
Ed., 2009, 48, 3322.
6 G. Ménard and D. W. Stephan, J. Am. Chem. Soc., 2010, 132,
1796.
7 A. Berkefeld, W. E. Piers and M. Parvez, J. Am. Chem. Soc.,
2010, 132, 10660.
8 A. Jana, D. Ghoshal, H. W. Roesky, I. Objartel, G. Schwab
and D. Stalke, J. Am. Chem. Soc., 2009, 131, 1288.
9 A. Jana, G. Tavčar, H. W. Roesky and M. John, Dalton
Trans., 2010, 39, 9487.
10 N. Takagi and S. Sakaki, J. Am. Chem. Soc., 2013, 135, 8955.
11 (a) L. W. Pineda, V. Jancik, K. Starke, R. B. Oswald and
H. W. Roesky, Angew. Chem., Int. Ed., 2006, 45, 2602;
(b) S. L. Choong, W. D. Woodul, C. Schenk, A. Stasch,
A. F. Richards and C. Jones, Organometallics, 2011, 30,
5543.
Conversion of 2 and 2′ to CH3OD with Me3N·AlH3
The germylene–formate 2 or 2′ (0.25 mmol) and Me3N·AlH3
(0.75 mmol) was placed in a Schlenk flask in the glovebox.
Toluene (10 mL) was added to the flask via a syringe at room
temperature under stirring. The mixture was allowed to stir for
another two hours, and cooled to 0 °C, D2O was added to the
solution and stirred for 10 minutes. The formed solid was sep-
arated using a centrifuge, and a clear two phase was formed.
The aqueous phase was collected and 1,4-dioxane was added
to it as an internal standard to determine the yields of CH3OD
by 1H NMR spectroscopy.9 Yields of CH3OD: 42% (2) and 46% (2′).
12 W. Wang, S. Inoue, S. Yao and M. Driess, Organometallics,
2011, 30, 6490.
Acknowledgements
13 W. D. Woodul, E. Carter, R. Müller, A. F. Richards,
A. Stasch, M. Kaupp, D. M. Murphy, M. Driess and
C. Jones, J. Am. Chem. Soc., 2011, 133, 10074.
14 In the latter case, the germylene hydride complex is
unstable, so compound 1 is the final product.
15 C. Cui, H. W. Roesky, H. Hao, H.-G. Schmidt and
M. Noltemeyer, Angew. Chem., Int. Ed., 2000, 39, 1815.
The authors thank the Cluster of Excellence UniCat for finan-
cial support (financed by the Deutsche Forschungsge-
meinschaft and administered by the TU Berlin).
Notes and references
1 (a) T. J. Marks, et al., Chem. Rev., 2001, 101, 953, (for a full 16 L. Ferro, P. B. Hitchcock, M. P. Coles and J. R. Fulton,
author list see ESI†) (b) Carbon Dioxide as Chemical Feed-
stock, ed. M. Aresta, Wiley-VCH, Weinheim, 2010.
2 P. G. Jessop, T. Ikariya and R. Noyori, Chem. Rev., 1995, 95,
259.
Inorg. Chem., 2012, 51, 1544.
17 Y. Yang, H. Zhu, H. W. Roesky, Z. Yang, G. Tan, H. Li,
M. John and R. Herbst-Irmer, Chem.–Eur. J., 2010, 16,
12530.
3 (a) F. Gassner and W. Leitner, J. Chem. Soc., Chem. 18 X. Ma, Z. Yang, X. Wang, H. W. Roesky, F. Wu and H. Zhu,
Commun., 1993, 19, 1465; (b) R. Tanaka, M. Yamashita and Inorg. Chem., 2011, 50, 2010.
K. Nozaki, J. Am. Chem. Soc., 2009, 131, 14168; (c) C. Ziebart, 19 G. M. Sheldrick, SHELX-97 Program for Crystal Structure
C. Federsel, P. Anbarasan, R. Jackstell, W. Baumann,
Determination, Universität Göttingen, Germany, 1997.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 6006–6011 | 6011